Ankle joint training method and system based on myoelectric signals

Abstract

The application relates to the technical field of rehabilitation training, and discloses an ankle joint training method and system based on myoelectric signals, which comprises the following steps: receiving myoelectric signals of a trainer and a selected training mode; when the training mode is an isokinetic training mode, adaptively adjusting a resistance torque applied to an ankle joint based on the myoelectric signals; and controlling an ankle joint rehabilitation robot to perform corresponding rehabilitation training actions according to the adjusted resistance torque. The embodiment of the application realizes the paradigm shift from mechanical programmed training to neural coupling personalized training of rehabilitation training in the isokinetic training mode based on myoelectric signals, and greatly improves the accuracy, safety and efficiency of rehabilitation training.

Description

Technical Field

[0001] This invention relates to the field of rehabilitation training technology, specifically to an ankle joint training method and system based on electromyographic signals. Background Technology

[0002] Ankle dysfunction is a common sequela of nerve injury, musculoskeletal injury, and in the elderly, severely affecting patients' walking ability and quality of life. Traditional ankle rehabilitation training often relies on manual operation by therapists or simple mechanical equipment, which has drawbacks such as difficulty in quantifying training intensity, strong subjectivity, and inability to accurately adapt to the patient's real-time muscle strength status.

[0003] While existing ankle rehabilitation robots can provide passive and active training modes, their resistance adjustment is mostly based on preset programs or simple kinematic parameters such as angle and speed, failing to delve into the core of neuromuscular function. This results in a lack of true biofeedback and adaptive capabilities during training, making it impossible to provide effective motor intention recognition and assistance for patients with weak muscle strength, and also making it difficult to achieve dynamic resistance challenges that match the patient's real-time muscle strength output during isokinetic training. Summary of the Invention

[0004] To address the aforementioned deficiencies, this invention discloses an ankle joint training method based on electromyographic signals, which can improve the accuracy, safety, and efficiency of rehabilitation training.

[0005] The first aspect of this invention discloses an ankle joint training method based on electromyographic signals, comprising:

[0006] Receives electromyographic signals from the trainee and the selected training mode;

[0007] When the training mode is isokinetic training mode, the resistance torque applied to the ankle joint is adaptively adjusted based on the electromyographic signals.

[0008] Based on the adjusted resistance torque, the ankle rehabilitation robot is controlled to perform corresponding rehabilitation training movements.

[0009] Based on electromyography signals, this invention realizes a paradigm shift in rehabilitation training under isokinetic training mode from mechanical, programmed training to neurally coupled, personalized training. This greatly improves the accuracy, safety, and efficiency of rehabilitation training and avoids the shortcomings of traditional rehabilitation equipment, which either have fixed resistance settings or are only adjusted to a limited extent based on simple motion parameters, making them completely unable to adapt to the patient's real-time changing muscle strength state.

[0010] This invention, through the introduction of electromyography (EMG) signals—a direct representation of neural drive—enables the system to perceive the trainee's intention and ability to exert force. For example, for a stroke patient with muscle strength only grade 2, when their EMG signals show enhanced neural drive but insufficient torque output, the system can intelligently reduce resistance and provide appropriate assistance to help them complete the movement, thereby establishing a neural pathway from motor intention to action execution. Conversely, for trainees with good muscle strength recovery, the system will automatically increase resistance based on strong EMG signals, creating an effective challenge.

[0011] As an optional implementation, in a first aspect of the present invention, adaptively adjusting the resistive torque applied to the ankle joint based on the electromyographic signal includes:

[0012] The basic torque is calculated based on the electromyographic signals and the actual angular velocity of the ankle rehabilitation robot:

[0013]

[0014] in, Basic torque, For feedforward gain parameters, This represents the electromyographic signal value. For feedback gain parameters, This is the actual angular velocity;

[0015] Based on the angular velocity error and its rate of change, the compensation torque is calculated using a fuzzy logic compensator:

[0016]

[0017] in, To compensate for the torque, For angular velocity error, Let F be the rate of change of angular velocity error, and F be the mapping function of the fuzzy logic compensator.

[0018] Calculate the resistance torque based on the aforementioned basic torque and compensation torque:

[0019]

[0020] in, This is the resistance torque.

[0021] This invention addresses the limitations of single control algorithms in handling nonlinear, time-varying physiological signals by combining MRAC (Model Reference Adaptive Control) and fuzzy logic compensation. The basic torque forms an adaptive master control loop with online self-learning capabilities, tracking the slow time-varying characteristics of the user's muscle strength. The fuzzy compensation torque handles transients and uncertainties. For example, if the user suddenly relaxes their exertion due to distraction, the fuzzy logic can quickly determine that this is a non-ability-related relaxation rather than fatigue, and immediately output a negative value. It performs rapid compensation to prevent speed loss; and when the user suddenly applies excessive force at a certain angle, it can smoothly apply suppression. This division of labor and cooperation makes the system response both fast and stable, effectively avoiding the oscillations and overshoot that may occur with pure MRAC, thus improving user experience and safety.

[0022] As an optional implementation, in a first aspect of the present invention, the feedforward gain parameter and the feedback gain parameter are characterized by time-varying adaptive parameters:

[0023]

[0024]

[0025] in, The rate of change of the feedforward gain parameter. For feedforward time-varying adaptive parameters, For the rate of change of the feedback gain parameter, To provide feedback for time-varying adaptive parameters, The correction factor is a positive constant.

[0026] This invention uses time-varying adaptive gain parameters to characterize the feedforward gain parameter and the feedback gain parameter, which can improve the stability and convergence of the control system. Time-varying adaptive gain and This allows the update rates of the feedforward and feedback gain parameters to be dynamically adjusted based on the system state, rather than being fixed. For example, during the initial training phase or when the direction of motion changes, if the error e is large, the system will automatically adopt a larger value. and Value, making and It converges quickly and adapts rapidly to the user's current capabilities; when approaching steady state, it adopts a smaller convergence rate. and The value is finely adjusted to avoid oscillations around the equilibrium point.

[0027] Correction factor This solves the parameter drift problem inherent in adaptive control. In real-world rehabilitation training, there may be continuous measurement noise or atypical, minute user movements. While these signals may not constitute an effective stimulus, they can still cause parameter drift in traditional MRAC. and Accumulate continuously until it diverges. (Introduction) and This is equivalent to providing a flexible regression force for the parameters, when the system lacks continuous excitation. and It will automatically and slowly shrink towards zero, rather than growing indefinitely, thus ensuring the long-term robust stability of the system.

[0028] As an optional implementation, in a first aspect of the present invention, the ankle rehabilitation robot is controlled to perform corresponding rehabilitation training movements based on the adjusted resistance torque, including:

[0029] The resistance torque is converted into a control current:

[0030]

[0031] in, To control the current, This is the torque coefficient. This is the resistance torque;

[0032] The control current is output to the actuator of the ankle rehabilitation robot.

[0033] Torque coefficient The torque coefficient can be obtained through experimentation. Different motors may have different torque coefficients. By converting the command torque into the input current (i.e., control current) to the actuator, a smooth torque output from the motor can be achieved, avoiding vibration or impact.

[0034] As an optional implementation, in the first aspect of the present invention, when the training mode is an isometric training mode, a guidance signal is generated using a PID algorithm based on the electromyographic signal error to guide the trainee in applying force to the ankle rehabilitation robot.

[0035]

[0036] in, As a guiding signal, For electromyography signal error, Let be a moment in the sampling period t. , , These are the proportional coefficient, integral coefficient, and differential coefficient, respectively.

[0037] This invention, specifically for isometric training, transforms the patient's invisible internal neural effort into externally visible and traceable guiding signals. For patients with nerve damage, the motor commands issued by their brains are often weak and uncoordinated. Traditional isometric training only encourages patients to push with all their might, lacking objective quantitative feedback; however, this invention calculates the error in electromyographic signals... The system uses a PID controller to generate guiding signals, providing patients with a clear target for exertion. For example, a cursor is displayed on the screen, and the patient needs to control their electromyographic signals to move the cursor to the target area. This training method can help patients rebuild precise neural control.

[0038] As an optional implementation, in the first aspect of the present invention, when the training mode is a traction training mode, the movement of the ankle rehabilitation robot is determined based on the comparison result of the electromyographic signal and a preset threshold.

[0039] When the electromyographic signal is less than or equal to a preset threshold, the ankle rehabilitation robot performs traction operation at the current speed;

[0040] When the electromyographic signal is greater than a preset threshold, one of the following methods shall be executed:

[0041] The ankle rehabilitation robot has stopped working.

[0042] Ankle rehabilitation robots reduce operating speed;

[0043] The ankle rehabilitation robot retracts by a predetermined safe angle in the opposite direction of traction, and continues to operate within a preset time when the electromyographic signal is less than or equal to a preset threshold.

[0044] This invention provides a safe, intelligent, and user-friendly interactive experience for traction training modes. It achieves intelligent protection through a three-level response mechanism: when the electromyography (EMG) signal value is less than or equal to a preset threshold, normal operation and continuous stretching continue; when the EMG signal value is greater than the preset threshold, the running speed is reduced or the angle is lowered to a safe level; when the EMG signal value is significantly greater than the preset threshold (e.g., the EMG signal value reaches three times the preset threshold), all movements are immediately and completely stopped. This gradient response strategy based on physiological feedback minimizes the risk of stretching injury without sacrificing rehabilitation effectiveness.

[0045] As an optional implementation, in the first aspect of the present invention, the method further includes:

[0046] When the absolute value of the rate of change of the resistance torque is greater than or equal to the preset upper limit of the rate of change, or / and the absolute value of the resistance torque is greater than or equal to the preset upper limit of the torque, the ankle joint rehabilitation robot is controlled to stop urgently.

[0047] This invention provides a hardware-level protection for rehabilitation training, independent of the control algorithm, by setting the resistance torque and its rate of change. For example, if the preset upper limit of torque is set to 40 N·m and the preset upper limit of the rate of change is 200 N·m / s, even if the control algorithm outputs a command of 50 N·m for some reason, or if the torque command changes by 3 N·m within 0.01 seconds due to a calculation jump (i.e., the rate of change of resistance torque reaches 300 N·m / s), the system will trigger an emergency shutdown. This fundamentally eliminates the possibility of mechanical injury to the user due to equipment overload or impact, and meets the safety standards for medical devices.

[0048] A second aspect of this invention discloses an ankle joint training system based on electromyographic signals, comprising:

[0049] The receiving unit is used to receive the trainee's electromyographic signals and selected training mode;

[0050] An adjustment unit is used to adaptively adjust the resistance torque applied to the ankle joint based on the electromyographic signals when the training mode is isokinetic training mode.

[0051] The control unit is used to control the ankle rehabilitation robot to perform corresponding rehabilitation training movements based on the adjusted resistance torque.

[0052] A third aspect of the present invention discloses an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calling the executable program code stored in the memory to execute the ankle joint training method based on electromyographic signals disclosed in the first aspect of the present invention.

[0053] A fourth aspect of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute the ankle joint training method based on electromyographic signals disclosed in the first aspect of the present invention. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is a schematic flowchart of the ankle joint training method based on electromyographic signals disclosed in an embodiment of the present invention;

[0056] Figure 2This is a schematic diagram of the process for generating mechanical cloud maps disclosed in an embodiment of the present invention;

[0057] Figure 3 This is a schematic diagram of the structure of an ankle joint training system based on electromyographic signals provided in an embodiment of the present invention;

[0058] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] It should be noted that the terms first, second, third, fourth, etc., in the specification and claims of this invention are used to distinguish different objects, not to describe a specific order. The terms used in the embodiments of this invention include and have, and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or devices.

[0061] Example 1

[0062] Please see Figure 1 , Figure 1 This is a flowchart illustrating the ankle joint training method based on electromyographic signals disclosed in an embodiment of the present invention. The execution entity of the method described in this embodiment is an execution entity composed of software and / or hardware. This execution entity can receive relevant information via wired or / or wireless means and can send certain instructions. It may also have certain processing and storage functions. This execution entity can control multiple devices, such as remote physical servers or cloud servers and related software, or local hosts or servers and related software that perform related operations on devices located in a certain location. In some scenarios, multiple storage devices can also be controlled; these storage devices may be placed in the same location as the devices or in different locations. Figure 1 As shown, this ankle joint training method based on electromyography signals includes the following steps:

[0063] S110 receives the trainee's electromyographic signals and selected training mode.

[0064] Electromyography (EMG) signals can be obtained using surface EMG sensors. These sensors employ high-density, differential electrode patches that can be installed on key ankle joint movement muscles such as the tibialis anterior, peroneus longus and brevis, and gastrocnemius muscles to accurately acquire ankle-related EMG signals. Understandably, the acquired EMG signals first need to be filtered by a bandpass filter to remove irrelevant low-frequency motion artifacts and high-frequency noise. The root mean square (RMS) or mean absolute value is then calculated as the EMG signal to characterize the degree of muscle exertion.

[0065] Training modes can be set on the ankle rehabilitation robot and selected by the trainer or therapist as needed. In a preferred embodiment of the invention, the training modes include three types: isokinetic training mode, isometric training mode, and traction training mode. Isokinetic training mode refers to maintaining a constant movement speed throughout the entire range of motion of the joint, requiring the trainer to exert maximum effort against resistance. Isometric training mode involves the trainer exerting static force at a specific angle, used for early muscle activation and muscle endurance training to achieve static muscle strength enhancement. Traction training mode is mainly for users with stiff joints and soft tissue adhesions, providing safe passive joint range of motion training.

[0066] In addition to electromyography (EMG) signals, the ankle rehabilitation robot is also equipped with an inertial measurement unit (IMU) and a pressure sensor. The IMU can capture the angle, angular velocity, and angular acceleration of the ankle joint in three-dimensional space in real time, accurately quantifying the range of motion and trajectory of the joint. The pressure sensor is placed on the foot pedal and is mainly used to monitor the distribution of plantar pressure and the applied torque in real time during isometric and isokinetic training modes.

[0067] The structure of the ankle rehabilitation robot is as follows Figure 2 As shown, it is a three-axis motion mechanism that can simulate the dorsiflexion / plantarflexion and inversion / eversion movement trajectory of a normal human ankle joint. It consists of three parts: robot body 210, foot pedal 220, and adjustable support frame 230. The output shaft of robot body 210 is fixedly connected to foot pedal 220. Foot pedal 220 and output shaft of robot body 210 form a pedal structure similar to a bicycle. Support frame 230 is located on one side of robot body 210 and is used to support the trainee's legs. It can be adapted to trainees with different lower limb sizes.

[0068] S120. When the training mode is isokinetic training mode, the resistance torque applied to the ankle joint is adaptively adjusted based on the electromyographic signal.

[0069] Employing a constant-speed training mode, the system dynamically and adaptively adjusts the resistance torque based on electromyographic (EMG) signals, enabling it to sense the trainee's intention and ability to exert force. For example, for a stroke patient with muscle strength only grade 2, when their EMG signals show enhanced neural drive but insufficient torque output, the system can intelligently reduce resistance, providing appropriate assistance to help them complete the movement, thereby establishing the neural pathway from motor intention to action execution. Conversely, for trainees with good muscle strength recovery, the system will automatically increase resistance based on strong EMG signals, creating an effective challenge.

[0070] In a preferred embodiment of the present invention, a combination of MRAC and fuzzy logic compensation is used to overcome the limitations of a single control algorithm in dealing with nonlinear, time-varying physiological signals. Specifically, the MRAC algorithm calculates the basic torque using real-time electromyographic signal values ​​and angular velocities (joint angles), thereby enabling the basic torque to form an adaptive master control loop with online self-learning capabilities, capable of tracking the slow time-varying characteristics of the user's muscle strength.

[0071]

[0072] in, Basic torque, For feedforward gain parameters, This represents the electromyographic signal value. For feedback gain parameters, This is the actual angular velocity.

[0073] It can be achieved through time-varying adaptive gain parameters and These parameters characterize the feedforward gain and feedback gain, thereby improving the stability and convergence of the control system.

[0074]

[0075]

[0076] in, The rate of change of the feedforward gain parameter. For feedforward time-varying adaptive parameters, For the rate of change of the feedback gain parameter, To provide feedback for time-varying adaptive parameters, The correction factor is a positive constant.

[0077] Time-varying adaptive gain parameter and This allows the update rates of the feedforward and feedback gain parameters to be dynamically adjusted based on the system state, rather than being fixed. For example, during the initial training phase or when the direction of motion changes, if the error e is large, the system will automatically adopt a larger value. and Value, making and It converges quickly and adapts rapidly to the user's current capabilities; when approaching steady state, it adopts a smaller convergence rate. and The value is finely adjusted to avoid oscillations around the equilibrium point.

[0078] And correction factor This solves the parameter drift problem inherent in adaptive control. In real-world rehabilitation training, there may be continuous measurement noise or atypical, minute user movements. While these signals may not constitute an effective stimulus, they can still cause parameter drift in traditional MRAC. and Accumulate continuously until it diverges. (Introduction) and This is equivalent to providing a flexible regression force for the parameters, when the system lacks continuous excitation. and It will automatically and slowly shrink towards zero, rather than growing indefinitely, thus ensuring the long-term robust stability of the system.

[0079] The fuzzy compensation torque can be calculated using a fuzzy logic compensator based on the angular velocity error and its rate of change.

[0080]

[0081] in, To compensate for the torque, For angular velocity error, Let F be the rate of change of angular velocity error, and F be the mapping function of the fuzzy logic compensator.

[0082] For example: the input variable is and , , , The target angular velocity is given by the output. .Will and The precise values ​​are converted into fuzzy linguistic values, such as {negative large, negative medium, negative small, zero, positive small, positive medium, positive large}, denoted as {NB, NM, NS, ZO, PS, PM, PB}; a fuzzy rule base is formulated based on expert experience, which is the mapping function F of the fuzzy logic compensator.

[0083] Fuzzy compensation torque is primarily responsible for handling transients and uncertainties. For example, when a user's concentration causes a sudden relaxation of effort, fuzzy logic can quickly determine that this is a non-capacity-related relaxation rather than fatigue, and thus immediately output a negative value. It provides rapid compensation to prevent speed loss; and it can smoothly suppress the force when the user suddenly exerts too much force at a certain angle.

[0084] Calculate the resistance torque based on the base torque and the compensation torque. :

[0085]

[0086] The division of labor and cooperation between MRAC and fuzzy logic compensation makes the system response both fast and stable, effectively avoiding the oscillations and overshoots that may occur with pure MRAC, thus improving user experience and security.

[0087] S130. Based on the adjusted resistance torque, control the ankle joint rehabilitation robot to perform corresponding rehabilitation training movements.

[0088] The resistance torque cannot be directly applied to the ankle rehabilitation robot; it needs to be converted into a control current and then output to the actuator of the ankle rehabilitation robot.

[0089]

[0090] in, To control the current, This is the torque coefficient. This is the resistance torque;

[0091] Torque coefficient The torque coefficient can be obtained through experimentation. Different motors may have different torque coefficients. By converting the command torque into the input current (i.e., control current) to the actuator, a smooth torque output from the motor can be achieved, avoiding vibration or impact.

[0092] For isometric training of patients with nerve damage, the motor commands issued by their brains are often weak and uncoordinated. Traditional isometric training can only make patients try their best to push, lacking objective quantitative feedback. In a preferred embodiment of the present invention, the invisible neural effort inside the patient can be transformed into externally visible and traceable guiding signals for isometric training.

[0093] Specifically, embodiments of the present invention calculate electromyographic signal errors. It also uses a PID controller to generate a guiding signal, providing the patient with a clear force target to guide the trainee in applying force to the ankle rehabilitation robot.

[0094]

[0095] in, As a guiding signal, For electromyography signal error, Let be a moment in the sampling period t. , , These are the proportional coefficient, integral coefficient, and differential coefficient, respectively.

[0096] It should be noted that the electromyographic signals in isometric training mode are normalized dimensionless data. For example, MVC can be used for normalization, so that the electromyographic signal value is between 0 and 1. The resulting guide signal is also dimensionless data between -1 and 1, which only represents the direction and intensity of force. For example, when the guide signal is negative (e.g., -0.8), it indicates overexertion and requires a command to reduce the force. The larger the absolute value, the stronger the command. When the guide signal is 0, no adjustment is needed. When the guide signal is positive (e.g., +0.6), it indicates underexertion and requires a command to increase the force. The larger the value, the stronger the command.

[0097] During implementation, the guiding signal can be converted into a cursor on the screen. The patient needs to control their electromyographic signals to move the cursor to the target area. This training method can help patients rebuild precise neural control.

[0098] This invention provides a safe, intelligent, and user-friendly interactive experience for traction training modes. It achieves intelligent protection through a three-level response mechanism: when the electromyography (EMG) signal value is less than or equal to a preset threshold, normal operation and continuous stretching continue; when the EMG signal value is greater than the preset threshold, the running speed is reduced or the angle is lowered to a safe level; when the EMG signal value is significantly greater than the preset threshold (e.g., the EMG signal value reaches three times the preset threshold), all movements are immediately and completely stopped. This gradient response strategy based on physiological feedback minimizes the risk of stretching injury without sacrificing rehabilitation effectiveness.

[0099] Specifically, when the training mode is traction training mode, the movement of the ankle joint rehabilitation robot is determined based on the comparison result of the electromyographic signal and the preset threshold.

[0100] When the electromyographic signal is less than or equal to a preset threshold, the ankle rehabilitation robot performs traction operation at the current speed;

[0101] When the electromyographic signal is greater than a preset threshold, one of the following methods shall be executed:

[0102] The ankle rehabilitation robot has stopped working.

[0103] Ankle rehabilitation robots reduce operating speed;

[0104] The ankle rehabilitation robot retracts by a predetermined safe angle in the opposite direction of traction, and continues to operate within a preset time when the electromyographic signal is less than or equal to a preset threshold.

[0105] Furthermore, to prevent abnormal movements of the rehabilitation robot from causing mechanical injury to the trainee, in a preferred embodiment of the present invention, a hardware-level protection independent of the control algorithm is provided for rehabilitation training by limiting the resistance torque and its rate of change. When the absolute value of the rate of change of the resistance torque is greater than or equal to a preset upper limit of the rate of change, or / and the absolute value of the resistance torque is greater than or equal to a preset upper limit of the torque, the ankle joint rehabilitation robot is controlled to stop urgently.

[0106] For example, the preset torque upper limit is set to 40 N·m and the preset change rate upper limit is 200 N·m / s. Even if the control algorithm outputs a command of 50 N·m for some reason, or if the torque command changes by 3 N·m within 0.01 seconds due to a calculation jump (i.e., the rate of change of the resistance torque reaches 300 N·m / s), the system will trigger an emergency shutdown. This fundamentally eliminates the possibility of mechanical injury to the user due to equipment overload or impact, and meets the safety standards for medical equipment.

[0107] Based on electromyography signals, this invention realizes a paradigm shift in rehabilitation training under isokinetic training mode from mechanical, programmed training to neurally coupled, personalized training. This greatly improves the accuracy, safety, and efficiency of rehabilitation training and avoids the shortcomings of traditional rehabilitation equipment, which either have fixed resistance settings or are only adjusted to a limited extent based on simple motion parameters, making them completely unable to adapt to the patient's real-time changing muscle strength state.

[0108] Example 2

[0109] Please see Figure 3 , Figure 3 This is a schematic diagram of the ankle joint training system based on electromyographic signals disclosed in an embodiment of the present invention. Figure 3 As shown, the ankle joint training system based on electromyography signals may include:

[0110] The receiving unit 310 is used to receive the electromyographic signals of the trainee and the selected training mode;

[0111] The adjustment unit 320 is used to adaptively adjust the resistance torque applied to the ankle joint based on the electromyographic signal when the training mode is isokinetic training mode.

[0112] The control unit 330 is used to control the ankle rehabilitation robot to perform corresponding rehabilitation training movements based on the adjusted resistance torque.

[0113] Preferably, the adjustment unit 320 may include:

[0114] The basic torque is calculated based on the electromyographic signals and the actual angular velocity of the ankle rehabilitation robot:

[0115]

[0116] in, Basic torque, For feedforward gain parameters, This represents the electromyographic signal value. For feedback gain parameters, This is the actual angular velocity;

[0117] The feedforward gain parameter and the feedback gain parameter can be characterized by time-varying adaptive parameters:

[0118]

[0119]

[0120] in, The rate of change of the feedforward gain parameter. For feedforward time-varying adaptive parameters, For the rate of change of the feedback gain parameter, To provide feedback for time-varying adaptive parameters, The correction factor is a positive constant.

[0121] Based on the angular velocity error and its rate of change, the compensation torque is calculated using a fuzzy logic compensator:

[0122]

[0123] in, To compensate for the torque, For angular velocity error, Let F be the rate of change of angular velocity error, and F be the mapping function of the fuzzy logic compensator.

[0124] Calculate the resistance torque based on the aforementioned basic torque and compensation torque:

[0125]

[0126] in, This is the resistance torque.

[0127] Preferably, the control unit 330 may include:

[0128] The resistance torque is converted into a control current:

[0129]

[0130] in, To control the current, This is the torque coefficient. This is the resistance torque;

[0131] The control current is output to the actuator of the ankle rehabilitation robot.

[0132] The system also includes an isometric training unit, which, when the training mode is isometric, uses a PID algorithm to generate a guidance signal based on the electromyographic signal error to guide the trainee in applying force to the ankle rehabilitation robot.

[0133]

[0134] in, As a guiding signal, For electromyography signal error, Let be a moment in the sampling period t. , , These are the proportional coefficient, integral coefficient, and differential coefficient, respectively.

[0135] The system also includes a traction training unit, which, when the training mode is traction training mode, determines the movement of the ankle rehabilitation robot based on the comparison result of the electromyographic signal and a preset threshold.

[0136] When the electromyographic signal is less than or equal to a preset threshold, the ankle rehabilitation robot performs traction operation at the current speed;

[0137] When the electromyographic signal is greater than a preset threshold, one of the following methods shall be executed:

[0138] The ankle rehabilitation robot has stopped working.

[0139] Ankle rehabilitation robots reduce operating speed;

[0140] The ankle rehabilitation robot retracts by a predetermined safe angle in the opposite direction of traction, and continues to operate within a preset time when the electromyographic signal is less than or equal to a preset threshold.

[0141] The system may also include a protection unit for controlling the ankle rehabilitation robot to stop urgently when the absolute value of the rate of change of the resistance torque is greater than or equal to a preset upper limit of the rate of change, or / and the absolute value of the resistance torque is greater than or equal to a preset upper limit of the torque.

[0142] Example 3

[0143] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. The electronic device can be a computer, a server, etc. Of course, in certain cases, it can also be a mobile phone, tablet computer, monitoring terminal, or other smart device, as well as an image acquisition device with processing capabilities. Figure 4As shown, the electronic device may include:

[0144] Memory 410 storing executable program code;

[0145] Processor 420 coupled to memory 410;

[0146] The processor 420 calls the executable program code stored in the memory 410 to execute some or all of the steps in the ankle joint training method based on electromyography signals in Embodiment 1.

[0147] This invention discloses a computer-readable storage medium storing a computer program that causes a computer to perform some or all of the steps in the ankle joint training method based on electromyography signals in Embodiment 1.

[0148] This invention also discloses a computer program product, wherein when the computer program product is run on a computer, the computer performs some or all of the steps in the ankle joint training method based on electromyographic signals in Embodiment 1.

[0149] This invention also discloses an application publishing platform, which is used to publish computer program products. When the computer program products are run on a computer, the computer performs some or all of the steps in the ankle joint training method based on electromyography signals in Embodiment 1.

[0150] In various embodiments of the present invention, it should be understood that the sequence number of each process does not necessarily imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0151] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they can be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0152] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0153] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several requests to cause a computer device (which can be a personal computer, server, or network device, specifically a processor in the computer device) to execute some or all of the steps of the methods described in the various embodiments of the present invention.

[0154] In the embodiments provided by this invention, it should be understood that B corresponding to A means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information.

[0155] Those skilled in the art will understand that some or all of the steps in the various methods of the embodiments described can be implemented by a program instructing related hardware. This program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0156] The above provides a detailed description of the ankle joint training method, system, electronic device, and storage medium based on electromyography signals disclosed in the embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An electromyographic signal-based ankle training system, characterized by, The system comprises: a receiving unit configured to receive an electromyography signal of a trainer and a selected training mode; an adjusting unit configured to, when the training mode is an isokinetic training mode, calculate a basic torque according to the electromyography signal and an actual angular velocity of an ankle rehabilitation robot; wherein is a base torque, is a feed forward gain parameter, is a myoelectric signal value, is a feedback gain parameter, is an actual angular velocity; calculate a compensation torque by using a fuzzy logic compensator according to an angular velocity error and a change rate of the angular velocity error; wherein, is a compensation torque, is an angular velocity error, is a rate of change of the angular velocity error, and F is a mapping function of the fuzzy logic compensator. calculate a resistance torque according to the basic torque and the compensation torque; wherein, Mdragis the drag moment; a control unit configured to control the ankle rehabilitation robot to perform a corresponding rehabilitation training action according to the adjusted resistance torque.

2. The myoelectric signal based ankle training system of claim 1, wherein, The feedforward gain parameter and the feedback gain parameter are characterized by time-varying adaptive parameters. wherein is a feedforward gain parameter rate of change, is a feedforward time-varying adaptive parameter, is a feedback gain parameter rate of change, is a feedback time-varying adaptive parameter, is a correction factor, and is a positive constant.

3. The myoelectric signal based ankle training system of claim 1 or 2, wherein, The control unit is configured to control the ankle rehabilitation robot to perform a corresponding rehabilitation training action according to the adjusted resistance torque, including: convert the resistance torque into a control current; wherein is the control current, is the torque coefficient; output the control current to an actuator of the ankle rehabilitation robot.

4. The myoelectric signal based ankle training system of claim 1, wherein, When the training mode is an isometric training mode, a guide signal is generated by using a PID algorithm according to the electromyography signal error, to guide the trainer to apply force to the ankle rehabilitation robot. wherein is a guidance signal, is an electromyographic signal error, is a time instant in a sampling period t, , , are a proportional coefficient, an integral coefficient and a derivative coefficient, respectively.

5. The myoelectric signal based ankle training system of claim 1, wherein, When the training mode is a traction training mode, an action of the ankle rehabilitation robot is determined according to a comparison result of the electromyography signal and a preset threshold value. When the electromyography signal is less than or equal to the preset threshold value, the ankle rehabilitation robot performs a traction operation at a current speed. When the electromyography signal is greater than the preset threshold value, any one of the following modes is executed: the ankle rehabilitation robot stops running; the ankle rehabilitation robot reduces a running speed; the ankle rehabilitation robot retreats by a predetermined safety angle in a direction opposite to the traction, and continues to run when the electromyography signal is less than or equal to the preset threshold value within a preset time.

6. The myoelectric signal based ankle training system of claim 1, wherein, The system further comprises a protection unit configured to control the ankle rehabilitation robot to stop running when an absolute value of a change rate of the resistance torque is greater than or equal to a preset upper limit of the change rate, or / and an absolute value of the resistance torque is greater than or equal to a preset upper limit of the torque.

7. An electronic device, comprising: The system comprises: a memory storing executable program codes; a processor coupled to the memory; the processor invokes the executable program codes stored in the memory, to perform: receiving an electromyography signal of a trainer and a selected training mode; when the training mode is an isokinetic training mode, calculating a basic torque according to the electromyography signal and an actual angular velocity of an ankle rehabilitation robot; wherein is a base torque, is a feed forward gain parameter, is a myoelectric signal value, is a feedback gain parameter, is an actual angular velocity; calculating a compensation torque by using a fuzzy logic compensator according to an angular velocity error and a change rate of the angular velocity error; wherein, is a compensation torque, is an angular velocity error, is a rate of change of the angular velocity error, F is a mapping function of the fuzzy logic compensator; calculating a resistance torque according to the basic torque and the compensation torque; wherein, Mdragis the drag moment; controlling the ankle rehabilitation robot to perform a corresponding rehabilitation training action according to the adjusted resistance torque.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, wherein the computer program causes a computer to perform: receiving an electromyography signal of a trainer and a selected training mode; when the training mode is an isokinetic training mode, calculating a basic torque according to the electromyography signal and an actual angular velocity of an ankle rehabilitation robot; wherein is a base torque, is a feed forward gain parameter, is a myoelectric signal value, is a feedback gain parameter, is an actual angular velocity; calculating a compensation torque by using a fuzzy logic compensator according to an angular velocity error and a change rate of the angular velocity error; wherein, is a compensation torque, is an angular velocity error, is a rate of change of the angular velocity error, and F is a mapping function of the fuzzy logic compensator. The resistance torque is calculated according to the base torque and the compensation torque: wherein Mdragis the drag moment; According to the adjusted resistance torque, the ankle rehabilitation robot is controlled to perform a corresponding rehabilitation training action.

Images